UC San Francisco is driven by the idea that when the best research, the best education and the best patient care converge, great breakthroughs are achieved. The Department of Anesthesia and Perioperative Care at UCSF is working hard to advance knowledge and yield scientific breakthroughs that will benefit life and health worldwide. Some of these efforts are listed below.
Elizabeth Whitlock, MD, MSc: NIH/NIA GEMSSTAR and FAER, “Impact of Coronary Revascularization on Longitudinal Cognitive Change in the Elderly”
This project will generate population – and individual-level data about cognitive change before and after heart procedures, which will help clinicians counsel older patients who need coronary revascularization and identify those at high risk of cognitive decline. Furthermore, this study will reconcile the research entity of “postoperative cognitive dysfunction” with clinical-relevant estimates of cognitive change and the medical necessity of these interventions in elders with cardiac disease, advancing the National Institute on Aging’s mission of promoting the health and well-being of older adults.
To view Liz’s profile, visit: https://profiles.ucsf.edu/elizabeth.whitlock
Bin Liu, PhD: NIH/NCI, “Novel Radioimmunoconjugates for Targeted Alpha-Particle Therapy of Metastatic Prostatic Cancer”
This project aims to develop novel antibody-based radioimmunotherapy against metastatic castration resistant prostate cancer.
To view Bin’s profile, visit: https://profiles.ucsf.edu/bin.liu
Christina Inglis-Arkell, MD: Mt. Zion Health Fund, “Preventing Opioid Over-Prescription in Post-Surgical Patients”
Co-Investigators: Solmaz Manuel, MD and Anjali Dixit, MD, MPH
This grant will build on previous quality improvement work which demonstrated over-prescription of postoperative opioids among patients at Mount Zion Surgery Center. They will work with surgical services to develop discharge guidelines for opioid prescriptions after surgery as well as implement preoperative patient interventions to maximize use of non-opioid analgesics after surgery.
To view Christy’s profile, visit: https://profiles.ucsf.edu/christy.inglis-arkell
To view Solmaz’s profile, visit: https://profiles.ucsf.edu/solmaz.manuel
To view Anjali’s profile, visit: https://profiles.ucsf.edu/anjali.dixit
Laura Lang, MD: ZSFG SURF, “Delivering Value-Based Care Through the Development of Enhanced Recovery After Surgery Pathways at ZSFG”
Co-Investigator: Benn Lancman, MBBS
The purpose of this project is the development of Enhanced Recovery After Surgery (ERAS) pathways at ZSFG to facilitate the delivery of value driven and evidenced based care. We aim to study our current perioperative environment, assemble a multidisciplinary core team, create infrastructure around pathway implementation and sustainability, and develop 3 distinct model cell pathways at ZSFG.
To view Laura’s profile, visit: https://profiles.ucsf.edu/laura.lang
To view Benn’s profile, visit: https://profiles.ucsf.edu/benn.lancman
David Shimabukuro, MD: Dascena subawards, NIH Prime:
R43 AA027674, “Early identification of acute kidney injury using deep recurrent neural nets, presented with probable etiology”
In this grant, a machine-learning-based prediction system is developed that continuously monitors for incipient AKI and offers clinicians probable root causes of AKI. Dr. Shimabukuro will provide his knowledge of AKI etiology and typical AKI presentation. Dr. Shimabukuro will also contribute his expertise working with UCSF data by selecting variables for the models that are available clinically in real time, by curating the algorithm adjustments so that they can continue to interface well with Epic and other Electronic Health Record systems, and by designing the system so that it integrates well with existing clinical workflows.
R43 HD096961, “Pediatric sepsis prediction: A machine learning solution for patient diversity”
In this grant, a machine-learning-based sepsis CDS system is developed by using multi-task learning techniques to define a set of “tasks,” each corresponding to predicting severe sepsis in a clinically distinct subpopulation of pediatric inpatients, together with a set of inter-task connections that share information between similar subpopulations. This will enable improved sepsis prediction across the widely heterogeneous and underserved pediatric patient population. Dr. David Shimabukuro will contribute his expertise in selecting variables for the models that are available clinically in real time, in curating the algorithm adjustments so that they can continue to interface well with Epic and other Electronic Health Record systems, and in designing the system so that it integrates well with existing clinical workflows. Further, he will contribute expertise in EHR integration of clinical decision support technology.
R43 TR002221, “A computational approach to early sepsis detection”
In this grant, transfer learning techniques are used to dramatically reduce the amount of target-site training data required by InSight, our machine-learning-based CDS tool for sepsis prediction, and empirically evaluate several such methods on a patient data set, using three different sepsis-related gold standards. This will enable InSight to deliver high performance, site-customized predictions despite access to only very small collections of target site training data, therefore allowing for the implementation of InSight at a wider range of settings than current pilot sites. Dr. Shimabukuro will contribute his expertise in selecting variables for the models that are available clinically in real time, in curating the algorithm adjustments so that they can continue to interface well with Epic and other Electronic Health Record systems, and in designing the system so that it integrates well with existing clinical workflows.
R43 TR002309, “Using clinical treatment data in a machine learning approach for sepsis detection”
In this grant, a “HindSight” software module is developed that recognizes sepsis, sepsis evaluation, and sepsis treatment. HindSight examines discharged patients’ electronic health records (EHR), identifies clinicians’ sepsis treatment decisions and patient outcomes, and passes these labeled examples to an online algorithm for the retraining of InSight, our machine-learning-based CDS tool for real-time sepsis prediction. Dr. David Shimabukuro and Dr. Mitchell Feldman at UCSF will assist with the clinical interpretation of study results and will ensure that results are statistically valid and meet the desired benchmarks, are reported accurately and are likely to be easily integrated into clinical workflows in Phase II.
To view David’s profile, visit: https://profiles.ucsf.edu/david.shimabukuro
Xiaobing Yu, MD: International Fibrodysplasia Ossificans Progressiva Association (IFOA) Accelerating Cures and Treatments (ACT): “Activin A/ACVR1 Modulation of Nociceptive Dysfunction in FOP”
Repeated painful flare-ups are the leading cause of emergency department visits and hospitalizations for FOP patients. Despite the prevalence and impact, our understanding of pain in FOP remains very limited. Using quantitative sensory testing (QST), we have recently discovered patients with FOP are often more sensitive to heat pain than healthy family members. To better understand the underlying mechanical and cellular mechanism, we will utilize patient iPSC-derived sensory neurons to study nociceptor dysfunction in FOP; and aim to identify more targeted pain therapies to improve patient care and comfort.
To view Xiaobing’s profile, visit: https://profiles.ucsf.edu/xiaobing.yu
Sen Gao, PhD: American Heart Association (AHA), “Somatic Mosaicism in Cerebral Arteriovenous Malformations”
Brain arteriovenous malformations (bAVMs) are tangled, abnormal blood vessels in the brain, that can cause symptoms varying form headache and seizures to life-threating stroke. Our research will focus on better understanding of how bAVMs develop, and why bAVM patients exhibit different degrees of disease severity and symptoms. We will identify factors contributing to bAVM characteristics and outcomes, especially stroke.
To view Sen’s profile, visit: https://profiles.ucsf.edu/sen.gao